sglang 0.4.6__py3-none-any.whl → 0.4.6.post2__py3-none-any.whl

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (85) hide show
  1. sglang/bench_one_batch.py +2 -0
  2. sglang/check_env.py +3 -3
  3. sglang/srt/configs/__init__.py +4 -0
  4. sglang/srt/configs/kimi_vl.py +38 -0
  5. sglang/srt/configs/kimi_vl_moonvit.py +32 -0
  6. sglang/srt/configs/model_config.py +15 -0
  7. sglang/srt/conversation.py +122 -1
  8. sglang/srt/disaggregation/decode.py +8 -2
  9. sglang/srt/disaggregation/fake/__init__.py +1 -0
  10. sglang/srt/disaggregation/fake/conn.py +88 -0
  11. sglang/srt/disaggregation/prefill.py +12 -3
  12. sglang/srt/disaggregation/utils.py +16 -2
  13. sglang/srt/entrypoints/engine.py +52 -21
  14. sglang/srt/entrypoints/http_server.py +27 -2
  15. sglang/srt/function_call_parser.py +97 -0
  16. sglang/srt/hf_transformers_utils.py +2 -0
  17. sglang/srt/layers/attention/cutlass_mla_backend.py +278 -0
  18. sglang/srt/layers/attention/flashinfer_backend.py +107 -82
  19. sglang/srt/layers/attention/flashinfer_mla_backend.py +27 -16
  20. sglang/srt/layers/attention/flashmla_backend.py +3 -0
  21. sglang/srt/layers/attention/utils.py +1 -1
  22. sglang/srt/layers/dp_attention.py +5 -2
  23. sglang/srt/layers/moe/ep_moe/token_dispatcher.py +1 -3
  24. sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=192,device_name=NVIDIA_A800-SXM4-80GB.json +146 -0
  25. sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=192,device_name=NVIDIA_H100_80GB_HBM3.json +146 -0
  26. sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=192,device_name=NVIDIA_H20.json +146 -0
  27. sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=192,device_name=NVIDIA_H200.json +146 -0
  28. sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=384,device_name=NVIDIA_H100_80GB_HBM3.json +146 -0
  29. sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=384,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json +146 -0
  30. sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=384,device_name=NVIDIA_H20.json +146 -0
  31. sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=384,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json +146 -0
  32. sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=384,device_name=NVIDIA_H200.json +146 -0
  33. sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=768,device_name=NVIDIA_A800-SXM4-80GB.json +146 -0
  34. sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=768,device_name=NVIDIA_H100_80GB_HBM3.json +146 -0
  35. sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=768,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json +146 -0
  36. sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=768,device_name=NVIDIA_H20.json +146 -0
  37. sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=768,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json +146 -0
  38. sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=768,device_name=NVIDIA_H200.json +146 -0
  39. sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=96,device_name=NVIDIA_H20.json +146 -0
  40. sglang/srt/layers/moe/fused_moe_triton/fused_moe.py +10 -8
  41. sglang/srt/layers/moe/fused_moe_triton/layer.py +15 -17
  42. sglang/srt/layers/quantization/__init__.py +2 -2
  43. sglang/srt/layers/quantization/deep_gemm.py +1 -1
  44. sglang/srt/layers/quantization/fp8.py +20 -22
  45. sglang/srt/layers/quantization/fp8_utils.py +2 -2
  46. sglang/srt/layers/utils.py +35 -0
  47. sglang/srt/lora/layers.py +35 -9
  48. sglang/srt/lora/lora_manager.py +84 -35
  49. sglang/srt/managers/data_parallel_controller.py +52 -34
  50. sglang/srt/managers/multimodal_processors/kimi_vl.py +73 -0
  51. sglang/srt/managers/schedule_batch.py +34 -15
  52. sglang/srt/managers/scheduler.py +273 -67
  53. sglang/srt/managers/scheduler_output_processor_mixin.py +26 -10
  54. sglang/srt/managers/tp_worker.py +52 -17
  55. sglang/srt/managers/tp_worker_overlap_thread.py +18 -7
  56. sglang/srt/mem_cache/memory_pool.py +70 -36
  57. sglang/srt/model_executor/cuda_graph_runner.py +82 -19
  58. sglang/srt/model_executor/forward_batch_info.py +31 -1
  59. sglang/srt/model_executor/model_runner.py +123 -58
  60. sglang/srt/models/deepseek_nextn.py +1 -257
  61. sglang/srt/models/deepseek_v2.py +78 -18
  62. sglang/srt/models/kimi_vl.py +308 -0
  63. sglang/srt/models/kimi_vl_moonvit.py +639 -0
  64. sglang/srt/models/llama.py +92 -30
  65. sglang/srt/models/llama4.py +2 -1
  66. sglang/srt/models/llama_eagle.py +4 -1
  67. sglang/srt/models/llama_eagle3.py +4 -1
  68. sglang/srt/models/qwen2_moe.py +8 -3
  69. sglang/srt/models/qwen2_vl.py +0 -12
  70. sglang/srt/models/qwen3_moe.py +8 -3
  71. sglang/srt/openai_api/adapter.py +49 -8
  72. sglang/srt/openai_api/protocol.py +13 -1
  73. sglang/srt/reasoning_parser.py +25 -1
  74. sglang/srt/server_args.py +83 -24
  75. sglang/srt/speculative/eagle_worker.py +3 -2
  76. sglang/srt/utils.py +91 -9
  77. sglang/test/runners.py +4 -0
  78. sglang/test/send_one.py +84 -28
  79. sglang/test/test_utils.py +67 -0
  80. sglang/version.py +1 -1
  81. {sglang-0.4.6.dist-info → sglang-0.4.6.post2.dist-info}/METADATA +5 -4
  82. {sglang-0.4.6.dist-info → sglang-0.4.6.post2.dist-info}/RECORD +85 -60
  83. {sglang-0.4.6.dist-info → sglang-0.4.6.post2.dist-info}/WHEEL +1 -1
  84. {sglang-0.4.6.dist-info → sglang-0.4.6.post2.dist-info}/licenses/LICENSE +0 -0
  85. {sglang-0.4.6.dist-info → sglang-0.4.6.post2.dist-info}/top_level.txt +0 -0
@@ -1,16 +1,16 @@
1
1
  sglang/__init__.py,sha256=T-fZEjKP66Q1q3PB56oREs5U3zf6bL0fNcdIbW8jMhE,1652
2
2
  sglang/api.py,sha256=vHiKBg8wwIdmrpnGclop5BzJ-1Q88emrlrfLwNCHg98,7010
3
3
  sglang/bench_offline_throughput.py,sha256=OQb-AjL4UNymmir02ht43uzgaNsnO_I11nXSowKMqBI,13841
4
- sglang/bench_one_batch.py,sha256=gUIYcFWM_oYSXnM4CHYJcyuX0l1aMG-afK7-iFjAJZI,19584
4
+ sglang/bench_one_batch.py,sha256=_xdxxdUqJtEGfURmBf0WFrQv38Q6vPlfG7QO2b4Fe6g,19622
5
5
  sglang/bench_one_batch_server.py,sha256=8VYNhaQbWGP8TkNVuy_sPjD5FiuVZHamtGRWKwa-Z-Q,5962
6
6
  sglang/bench_serving.py,sha256=8rbek9PLYEHdt8fdll-z_P9e6GpmlLohHiyqY99JXIs,57567
7
- sglang/check_env.py,sha256=76itNLUw9KlqbiY1BI4u4YaMZaqyCNcrCLUIb6aHflM,8396
7
+ sglang/check_env.py,sha256=qDMIG2rCNBH1yKnxQmF-Bp10oiFMUKMgfZLHZYOmdSY,8412
8
8
  sglang/compile_deep_gemm.py,sha256=Umy3oYFeCn40qHUdwlPVuFXmA24fFYB-fuWApgZnEfw,6211
9
9
  sglang/global_config.py,sha256=xzLdk8W53fneFblNh8iIjGF9C3-7mnzR1-LleD9Btxg,1495
10
10
  sglang/launch_server.py,sha256=mDXfwha8LHpWQJekcCosR98QhCQsbmilsBlI5jAIgg0,420
11
11
  sglang/llama3_eval.py,sha256=gWSboDchIGybIce88bJlrCG0yiLZ513mw4gcutJlzGM,10017
12
12
  sglang/utils.py,sha256=GIcgiRHkZ-gyPxXOdn1qFF41jkg4-YdDxbPc4mzO-qk,16159
13
- sglang/version.py,sha256=bbBpXE_PBbo_SaI807mDML0QJywD0_ufCDPgAMlDHaE,22
13
+ sglang/version.py,sha256=EOm4637qsQ4gxuzOdaxFFBpWIBeQl7iFLSjMjGurEHg,28
14
14
  sglang/lang/chat_template.py,sha256=MwNL5dNTe8g_l2ljZubnrazEgT2xEv-9O2D0Ezwxy4I,19658
15
15
  sglang/lang/choices.py,sha256=-W1DVw9N9ZliVpvmWrzIXG4cswAah8eMQrHWzkS3D8o,6234
16
16
  sglang/lang/compiler.py,sha256=MAuzoOOpb98njJ7Io2SDmFkhTroDYiq0te0ZpfHkMY4,7597
@@ -27,27 +27,29 @@ sglang/lang/backend/vertexai.py,sha256=gz0uNYyBb88jbPYz6ZIJ774fefrcbuVdoK33bphUZ
27
27
  sglang/srt/_custom_ops.py,sha256=L7NuEaRD_Q6Q54n0NZnLXgWZURbnn8Tkg4NQedE6zgA,3616
28
28
  sglang/srt/aio_rwlock.py,sha256=6LYtOdeTUY3hkfa1dmYkgsaF2ttrwIF3hUWz2AZ2fqw,2970
29
29
  sglang/srt/code_completion_parser.py,sha256=iYRFBxXBAysHcBnf9IHmmyjVkrqKu_9h6Z0_EEjjTp4,5404
30
- sglang/srt/conversation.py,sha256=jgm15yl2SPjSlVjLPwWYklUsUUElq-7W6-KqqGc30vs,30262
30
+ sglang/srt/conversation.py,sha256=otEs_H81NcYsy1PKx8l5Q0BUOIXjHscBlRVEDApWQos,34020
31
31
  sglang/srt/custom_op.py,sha256=J1PUcGaeJJjfAjp06BQsLpUkKyR1zsh9MvDiDlqqJsg,1129
32
- sglang/srt/function_call_parser.py,sha256=gkCzjf7F2xYUmRunrOKzuB_biTdTKxdA1Vil-v2NlCs,29546
33
- sglang/srt/hf_transformers_utils.py,sha256=N2f-gA8yUq-UP_TJT276gNbDNzmddWsmWnq3px6TIj8,9342
32
+ sglang/srt/function_call_parser.py,sha256=evwCPbLFjgNiRf-1CcxVFYbF08UVsh0ZGvq_R35TPlw,33379
33
+ sglang/srt/hf_transformers_utils.py,sha256=JEDiaMYzq89sXWIu_CD2kmnNF1OA-gUJOkyh6kWolr0,9403
34
34
  sglang/srt/mm_utils.py,sha256=1ScBunw_x4W8ebM_AcJ62-1T2mfT8NlMJqdAhkF1lb0,12367
35
35
  sglang/srt/model_parallel.py,sha256=eLXZhvJ4wG6dh0FontNCIdVZvHYdWgaeY-5cu7TD9tE,6078
36
36
  sglang/srt/patch_torch.py,sha256=OUPCGQSQz3MVZB1zZ_Eq8lXiw0uIKJ_HWjqQolI8FsM,3088
37
- sglang/srt/reasoning_parser.py,sha256=JnaEVW0KG1yJpn9uxmrjwErb9imzni05QDFjExryoqM,5584
38
- sglang/srt/server_args.py,sha256=ojKNFeiZ1W7nQScImdlG4xuHVR1yQdugwrlHj30WVK8,54804
37
+ sglang/srt/reasoning_parser.py,sha256=8AMk3XI8mfvz4AUuRHf_pNYpM_Mr64uT9EZ3o90cqQ8,6341
38
+ sglang/srt/server_args.py,sha256=I4Gxme7w3Fpzj6tsbGu8LmLaMLpxnhqqFXSGVjh6Pi8,56945
39
39
  sglang/srt/torch_memory_saver_adapter.py,sha256=HYlgYJ2sgmjs2RSjU2KbCaXijRg3mTDZ0ZcCB5Bt6Ps,2211
40
- sglang/srt/utils.py,sha256=FsLjbxNf1K1hmQTTpQ7PyZxze0m1nfyKfNhngfC8atU,62944
40
+ sglang/srt/utils.py,sha256=AtpND4l1ho_h5N6BIVBRgPAIKjG4DomqRQa_Z08ugvs,65335
41
41
  sglang/srt/warmup.py,sha256=FmJiYfjRr3X_eAe7ojQaPoN17LvHpjDmRWRnO-k86AQ,1469
42
- sglang/srt/configs/__init__.py,sha256=vulncVn70WqIT6s0HaB8p_Q6FjOiaLwNZWpoJS9FIuQ,399
42
+ sglang/srt/configs/__init__.py,sha256=8EcVRP95epZ49DxBa6LgKWt7eO3Qe7Hrr3V1c6HkMnY,553
43
43
  sglang/srt/configs/chatglm.py,sha256=j-b0YkdYUmQm2y1kNmMJtKeACxWKmBbvNNkDWbs6kbI,2907
44
44
  sglang/srt/configs/dbrx.py,sha256=tdhIkXAQl1yr0MxqFmsDG1E0e2puRTTKm6UTyANBLac,11005
45
45
  sglang/srt/configs/deepseekvl2.py,sha256=21jZravchHcwyTQ5ROu1NzwI_eu-ngt3v8SRMm4XE0k,23327
46
46
  sglang/srt/configs/device_config.py,sha256=kfmpPOECqYxcRoY-ko0QZRhyiBWUGP2CMF51DMUN5nU,435
47
47
  sglang/srt/configs/exaone.py,sha256=Duxd4yQoKy8GWEzZD_kCY_OzmN_67CTJL_Kgn0eXk3g,10731
48
48
  sglang/srt/configs/janus_pro.py,sha256=-QtJ4ZGZiAJb0AkOEcuCHzIKLw23nF8nRk3rdCcoUO0,19016
49
+ sglang/srt/configs/kimi_vl.py,sha256=4W7VQI3pr888ZsFA2SqCQo4mI0seXTOrGQ-x3oTvWew,1358
50
+ sglang/srt/configs/kimi_vl_moonvit.py,sha256=hx2Rt4JSFbvy2HUTeLjBpge87m8M6ITAhqsgdNf_Jd4,1163
49
51
  sglang/srt/configs/load_config.py,sha256=qs-AxuplouBx2tsv9KGBOLZPbwzuVA4vbktbGP_cRp8,3309
50
- sglang/srt/configs/model_config.py,sha256=GtVEAqxcitVldxLroaHYwoILjfa--a2KmbcBMyyeF08,22421
52
+ sglang/srt/configs/model_config.py,sha256=xyVgImD3bvaDuDuGnWsB-Z81McWzs83Sju3M7J54f8A,23134
51
53
  sglang/srt/configs/utils.py,sha256=3nHUfisMs_Ltuhv8OZTNCJp63YJKJVF43h1QZB1zqx8,670
52
54
  sglang/srt/connector/__init__.py,sha256=czLX5JOxuMhH-T9eSJzoc1qv1B4z9chyffDRL5I6wo4,1247
53
55
  sglang/srt/connector/base_connector.py,sha256=i6i1TIzsz4NbSEkrdMPq-urb2sN2aLAx8dazga4gB9U,2833
@@ -64,12 +66,14 @@ sglang/srt/constrained/outlines_jump_forward.py,sha256=Gyubp-FVetxd6wP4FA_kD6cCX
64
66
  sglang/srt/constrained/reasoner_grammar_backend.py,sha256=XFxdZqvPofmtCeIMqR10NOyph06HwbdXfiVI8rIoV5s,3646
65
67
  sglang/srt/constrained/xgrammar_backend.py,sha256=oc3BTTe8mB5Szv_O-5nZzWbKEKFb22oUniqTUZhewLQ,7409
66
68
  sglang/srt/constrained/triton_ops/bitmask_ops.py,sha256=WjTen9iuuFWLzkE1mAHQZB9_7aIy5QH8Wjf-lB-Fams,4614
67
- sglang/srt/disaggregation/decode.py,sha256=G9iYiCGEyI-bFFuLQhgsEN75bzhYtb--QCvlzbbSm1o,24884
69
+ sglang/srt/disaggregation/decode.py,sha256=nSHCBfEtD3a6c2a7XPAcCh4c0jw3BLG9EL-L3LlW_V0,25139
68
70
  sglang/srt/disaggregation/mini_lb.py,sha256=zyJo20GI6G1ZIoBVY3ltcr0dDcH5qOJrtMfiGKGnBLI,10959
69
- sglang/srt/disaggregation/prefill.py,sha256=VE0t6SasG_4QNeEpBkNbwicEpR-SyEsyYSMPROERH2k,15100
70
- sglang/srt/disaggregation/utils.py,sha256=6ackrcitNF94ur9bRvPerF29E7g8uU_0a4-nSdw1EMs,3078
71
+ sglang/srt/disaggregation/prefill.py,sha256=4wLYQtPMbKWMQvF3mGnvr8ygd9xRLO9zTwLKeM5BIf8,15424
72
+ sglang/srt/disaggregation/utils.py,sha256=7gO734GOr4u03qwOf2UvFsfj4n-I37iyzQh7lFKbJRk,3501
71
73
  sglang/srt/disaggregation/base/__init__.py,sha256=KR8xXoRCDAy2U623mfP6ujXu42m1_F9EiudjrKu2I_A,130
72
74
  sglang/srt/disaggregation/base/conn.py,sha256=gpf32bhYXWm_iaYB6WcrDaJ-UoL1ZzPI_xpi5pMhRQo,2443
75
+ sglang/srt/disaggregation/fake/__init__.py,sha256=zmfeKYXjonRhfFOck1c_mP7Q4cW5G0f1RsTwRivKu0s,47
76
+ sglang/srt/disaggregation/fake/conn.py,sha256=DKEVBgmzV3CNzZ0-r7rFV4orue2iP_7apEtgn-fcTEA,2552
73
77
  sglang/srt/disaggregation/mooncake/__init__.py,sha256=1vacEHmWjf7zgbMPzsXKB08FqNKNCquJdUiDlO41BOk,122
74
78
  sglang/srt/disaggregation/mooncake/conn.py,sha256=DQ_PTxq_nZHFZ4LxHDhCIvQFPA1xUbaw1Sleyqhkq6U,28224
75
79
  sglang/srt/disaggregation/mooncake/transfer_engine.py,sha256=MxDAB9ZetRF1pFS2LP3FVHPtQ1HjIt_SK3UMaYHZ94o,2604
@@ -88,12 +92,12 @@ sglang/srt/distributed/device_communicators/pynccl_wrapper.py,sha256=LblisImY9d6
88
92
  sglang/srt/distributed/device_communicators/shm_broadcast.py,sha256=bbruDIM1GgKIdB6gi71_I0mpB179I-qyvwKuSj1Kaic,20816
89
93
  sglang/srt/distributed/device_communicators/xpu_communicator.py,sha256=ajW6132BvA6jkeipEIgN27TFycI0U06Ih2Z8WNjlA4s,1593
90
94
  sglang/srt/entrypoints/EngineBase.py,sha256=xoyvp6XAeDLY2_Q2Ng33H-fRhrXHv2ldJJKd-HuDhqE,1870
91
- sglang/srt/entrypoints/engine.py,sha256=-qHfYFDremLj8GhwQa7GuWHQKop7Tyq2HvmBXULtCos,22282
92
- sglang/srt/entrypoints/http_server.py,sha256=4fbn2KfUs2fjSvDulnhGfcEGA5a9ahUEre59TazwVA4,28508
95
+ sglang/srt/entrypoints/engine.py,sha256=4S0ubmNIo8adkapEUOram1MXjkmyiBWXVoHCjXyHAIw,23430
96
+ sglang/srt/entrypoints/http_server.py,sha256=vvyvCosUp5aTFlD8k4IyZDzj2yXQIsndhPkTl4u1nGc,29573
93
97
  sglang/srt/entrypoints/http_server_engine.py,sha256=ihA6y3GXRs28Y9U3SgdQcJQjnw_SVIby7QrVgiafX04,4846
94
98
  sglang/srt/entrypoints/verl_engine.py,sha256=XLYdwTwhH0jTjw8xczgZXWfBXMRb_ur2bg4TN0dTwfI,6975
95
99
  sglang/srt/layers/activation.py,sha256=oSkdo8B8najXFcVay3Y__CEvgXh87lAIhG0CMp2Ugqs,5954
96
- sglang/srt/layers/dp_attention.py,sha256=WJgXg_KyBzDHkwyfUFBowpDRFd0q5Q9LgEhqT-qT_ys,7549
100
+ sglang/srt/layers/dp_attention.py,sha256=I5cJnBT996mzjpNRrzcZXGt9j8nrkgD4A4T-BHiHkGM,7649
97
101
  sglang/srt/layers/elementwise.py,sha256=XCrR2i-9dP-H6jQo2zUuquwZrsl_wEQqj5Wxk6WUf7o,13987
98
102
  sglang/srt/layers/layernorm.py,sha256=2XEaRK9e6syWO3YLcqWqlR7hZ5R-CFDqbCII-zntQLM,5957
99
103
  sglang/srt/layers/linear.py,sha256=nC9MxJrFap1BEyqgFlBySH4IeQruIbcBp32cOhUl5Fw,52149
@@ -104,16 +108,18 @@ sglang/srt/layers/radix_attention.py,sha256=F71GgDes_fEt_cHxR9HM2QhNG5u7uF4zDAuL
104
108
  sglang/srt/layers/rotary_embedding.py,sha256=eVBwYvGpFhL1KyyPutQuZotmvSpChcxzyhpmcbQ6cKQ,48267
105
109
  sglang/srt/layers/sampler.py,sha256=PNgMXm2vsNsE6Rt89R5GLDC3lDxdIujoWli8F3vldng,11384
106
110
  sglang/srt/layers/torchao_utils.py,sha256=Ws24FdRBSkTpyeyA6bQrdDm-W5wfDxKvSIPUSahyMfA,4063
111
+ sglang/srt/layers/utils.py,sha256=tkTz86DFZ4NRMEUc4QkYNskUskdxXoEqkWqaMGbhP7E,1045
107
112
  sglang/srt/layers/vocab_parallel_embedding.py,sha256=QUxd4sELx6p3dHvEKmccPZ-phdd_9EjNdwjH3SJ9zxI,22238
108
113
  sglang/srt/layers/attention/base_attn_backend.py,sha256=lGujcYJ_CxHJy0Q9L6Phn3ds-nBGMy0OGj3R54R65iQ,3334
114
+ sglang/srt/layers/attention/cutlass_mla_backend.py,sha256=kCNn5Ub0jCsFjhVhuE-9qK53gu5a9oMStMeT2lcc6CU,9904
109
115
  sglang/srt/layers/attention/double_sparsity_backend.py,sha256=2ZRL_gYz14idoVqQzeQ6N77nXer0f_8_TUYw40XUUz0,9161
110
116
  sglang/srt/layers/attention/flashattention_backend.py,sha256=ysJt9pJ8pg_kVxvVUTvUL22-O7ABHCenLGGcqCotD6A,83206
111
- sglang/srt/layers/attention/flashinfer_backend.py,sha256=YtMTgMhxxNrAbSoWTPJczgY4SR3WjnAPXPoJ2d5PUZY,46394
112
- sglang/srt/layers/attention/flashinfer_mla_backend.py,sha256=pnVhvVEK87iFW8gUb1G7X7c1tqro8R2DSEOFCnlV8Bo,30301
113
- sglang/srt/layers/attention/flashmla_backend.py,sha256=IyE4w7GcNOxjjy3mQeuAMjPtBNvI-6JkoxvBlFxFvec,10270
117
+ sglang/srt/layers/attention/flashinfer_backend.py,sha256=4TavTj9AH4aVccsnDbulYnJ0hKcHgWm9gpNtcbsf2kI,47099
118
+ sglang/srt/layers/attention/flashinfer_mla_backend.py,sha256=LBMAeC1OFEQ6hjN1GHpaURrM-yWKC_lxTqqERvvehJ8,30616
119
+ sglang/srt/layers/attention/flashmla_backend.py,sha256=wR0bkLz3mj5EfuHEi9fwEP2vtq6xFhsrIijpFb009o4,10340
114
120
  sglang/srt/layers/attention/torch_native_backend.py,sha256=K5hUqBgakk2COSQqsaxWs0yEVOHS-7BlOygZTOeI8kE,9444
115
121
  sglang/srt/layers/attention/triton_backend.py,sha256=oEEiUwHbm4rNw5ExbQ2c3n0TwAgkk77yuLFenj9bHOo,26902
116
- sglang/srt/layers/attention/utils.py,sha256=J9mA-cbZT3uTlaKXo0HEAaeMei_TS2o4McTna9LVDCE,2750
122
+ sglang/srt/layers/attention/utils.py,sha256=rxB4sbNIHDTges78bDbnpd_hUgtyb3e16wUwgI4WmoU,2751
117
123
  sglang/srt/layers/attention/vision.py,sha256=CtFU1wyz5191LcuyDzGJ01mB-mM-upPj2pXg4DO6wh4,11985
118
124
  sglang/srt/layers/attention/triton_ops/decode_attention.py,sha256=DPu_aCPgwPqKWZPEQmp_xA7MPbpV2ip-MEICCB470Ao,19120
119
125
  sglang/srt/layers/attention/triton_ops/double_sparsity_attention.py,sha256=vsDZZ5QGb8-KBzADgKshnVQbsW8zRJF1h5hgdPGW5lU,31124
@@ -126,10 +132,10 @@ sglang/srt/layers/moe/topk.py,sha256=K-VU64nWBV07bu1Okn-uYbhz9gylq-KFNRYn2SFzu28
126
132
  sglang/srt/layers/moe/ep_moe/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
127
133
  sglang/srt/layers/moe/ep_moe/kernels.py,sha256=ijqRzS-tb0LGnDU5hW-g0JH104ppADrWaUIDGxb9Feo,22919
128
134
  sglang/srt/layers/moe/ep_moe/layer.py,sha256=SZ0shPwgDp7xj-TCv9bfg5O7f2AXjF6xmBP5xkZ0Ips,36440
129
- sglang/srt/layers/moe/ep_moe/token_dispatcher.py,sha256=zQV7Qr-Zrcr3D3efVvZepRQM02bj5djHPsijPssavk8,20430
135
+ sglang/srt/layers/moe/ep_moe/token_dispatcher.py,sha256=zqZCOsBk-fNRQHxmgt8XSgUG6AJjdKFRgJZ4YtuItJ4,20383
130
136
  sglang/srt/layers/moe/fused_moe_triton/__init__.py,sha256=h9yMFAL_bagUf-qBED8gSWdCOb7d8IdA-pE-L_nIg8E,842
131
- sglang/srt/layers/moe/fused_moe_triton/fused_moe.py,sha256=bgKelgR3IPgdK4ihasWL4EaFi_z4mVnOKDeMLAio8hc,55951
132
- sglang/srt/layers/moe/fused_moe_triton/layer.py,sha256=XofcceHnkmqhVczEIVqGSbsxps5LiYTGcaCk1Say9YM,24951
137
+ sglang/srt/layers/moe/fused_moe_triton/fused_moe.py,sha256=HJerp49tivE-IBZpy9JQs0UMx4goPGGO6gU6RINe2rk,55930
138
+ sglang/srt/layers/moe/fused_moe_triton/layer.py,sha256=BMOV76fabrZcoyDmRpRbH11Jc0ogWH2k2QAQwvZIpgI,25084
133
139
  "sglang/srt/layers/moe/fused_moe_triton/configs/E=1,N=14336,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json",sha256=iNGsE2ZeVnQEnN4A8UJ9Jv0d3hbRF2MJ9oBgjup5Szk,2737
134
140
  "sglang/srt/layers/moe/fused_moe_triton/configs/E=1,N=14336,device_name=NVIDIA_A100-SXM4-80GB.json",sha256=JJN0hryyLr5Zv3dSS7C8cPFhAwTT6XxUVnBGMZvV6JA,2752
135
141
  "sglang/srt/layers/moe/fused_moe_triton/configs/E=1,N=1792,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json",sha256=ouRyZ5PEMPP2njPftCNhs-1g1y6wueWLmhI7G1SjV1k,4131
@@ -141,7 +147,23 @@ sglang/srt/layers/moe/fused_moe_triton/layer.py,sha256=XofcceHnkmqhVczEIVqGSbsxp
141
147
  "sglang/srt/layers/moe/fused_moe_triton/configs/E=1,N=3584,device_name=NVIDIA_A100-SXM4-80GB.json",sha256=yf33YmWlVSjjyg0Q4OMAWvc9gjRxvttMrQBUEOfPl4I,4153
142
148
  "sglang/srt/layers/moe/fused_moe_triton/configs/E=1,N=7168,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json",sha256=ZWMClYN1moVRUP2f0hYac38di_pUgZggyl9d2D5rnoc,4136
143
149
  "sglang/srt/layers/moe/fused_moe_triton/configs/E=1,N=7168,device_name=NVIDIA_A100-SXM4-80GB.json",sha256=C65Q2Mv1LxFQ_qDnv11IZ9nwl7sGZo72nWDflMttu4g,4147
150
+ "sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=192,device_name=NVIDIA_A800-SXM4-80GB.json",sha256=T-_T-oW4qpjTIBaGVxukJksRE7Yg8m9HNHgJ2XmR3aI,3242
151
+ "sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=192,device_name=NVIDIA_H100_80GB_HBM3.json",sha256=I3k416HbXU_rYb8scD8gAI4fuBlElHl06PM347Qa11w,3253
152
+ "sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=192,device_name=NVIDIA_H20.json",sha256=RgV8C4F1LO09h01YsgF_eqX6GNoBtC7ulPfJRUUbg_g,3241
153
+ "sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=192,device_name=NVIDIA_H200.json",sha256=nsNEuDNks0tVLfQfIm7xxFwEeptTfQcoa9fJy0NS8xQ,3247
154
+ "sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=384,device_name=NVIDIA_H100_80GB_HBM3.json",sha256=R4gBc3sMY5QwOtcGwGKdk2Ak4UsUbBd3jDUeKKk0O1U,3257
155
+ "sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=384,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=qbqjisJ4oKmcYzumHPRk5UyOzsdi8J6xas82UWHMeAI,3263
156
+ "sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=384,device_name=NVIDIA_H20.json",sha256=vS2DRIDOqWyiBvbG6H746ownfkD1F8Aj2YZ0ET9xll8,3232
157
+ "sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=384,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=1n5XyZZ5sKAi-Z1duWOhLUfr6gkvnOpvxfbqIT6iU_4,3265
158
+ "sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=384,device_name=NVIDIA_H200.json",sha256=xqhl748it8GV2KXX0XixitE_ywnsKksqK8AGL7tAgT8,3254
144
159
  "sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=512,device_name=NVIDIA_H100_80GB_HBM3.json",sha256=FsWbV4Q6AzAtgegVuENBDz2ZcSJsqNiwUIVfQbpP7hQ,3244
160
+ "sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=768,device_name=NVIDIA_A800-SXM4-80GB.json",sha256=T5rXJOZYNEs_3hE8g3ch802DnySbNiIqdn0s0RlJr8U,3249
161
+ "sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=768,device_name=NVIDIA_H100_80GB_HBM3.json",sha256=9L5C8VcSsiUr5XryXB1AO3DknlAQowp6DU6S7OSzEA0,3248
162
+ "sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=768,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=IuvyC8TNhCVAmUZfLSoETsyCKsmejKXrs_0zuwFLPAU,3265
163
+ "sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=768,device_name=NVIDIA_H20.json",sha256=10Ntu2aVD5vGLonx-jW0qNw-tgZWdZmzMGx7utDVeng,3237
164
+ "sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=768,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=pdQ1RvXvdWDn8Y8-8MAX3vn-T-wbtkZvHV9GZZvNjnc,3266
165
+ "sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=768,device_name=NVIDIA_H200.json",sha256=JraM-Nvbg5V_TJkSl6UPFYZN1zHHoIbr2pAcksenoTY,3248
166
+ "sglang/srt/layers/moe/fused_moe_triton/configs/E=128,N=96,device_name=NVIDIA_H20.json",sha256=JtcHRlPz8xQEAqJ9EWI63oYvdmjQFG6VTHqtt85VOSA,3221
145
167
  "sglang/srt/layers/moe/fused_moe_triton/configs/E=144,N=512,device_name=NVIDIA_H100_80GB_HBM3.json",sha256=pk6VZChh2Y0CsJSzjtUhOnlta1QLTUEWy33aKQU47XY,3244
146
168
  "sglang/srt/layers/moe/fused_moe_triton/configs/E=16,N=1024,device_name=NVIDIA_H100_80GB_HBM3.json",sha256=Gmk24hc5lVIfQtqSa5wLOcWKedMN8aZUe93DBh6J1AY,3249
147
169
  "sglang/srt/layers/moe/fused_moe_triton/configs/E=16,N=1024,device_name=NVIDIA_H200.json",sha256=uY_XMPomaXMXxIkTR4ctU_Ybri_jMv2VvCcV-f6O_bw,3255
@@ -265,14 +287,14 @@ sglang/srt/layers/moe/fused_moe_triton/layer.py,sha256=XofcceHnkmqhVczEIVqGSbsxp
265
287
  "sglang/srt/layers/moe/fused_moe_triton/configs/E=8,N=8192,device_name=AMD_Radeon_Graphics,dtype=fp8_w8a8.json",sha256=-RzUWSIAAsg6iA-8SPMa68hPpBVoUyMJs3dLP7edRu0,4323
266
288
  "sglang/srt/layers/moe/fused_moe_triton/configs/E=8,N=8192,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8.json",sha256=sY2nWMPh9lsIkhPCjkHO245wpnfFbrHmzdcZDVFPVww,3265
267
289
  "sglang/srt/layers/moe/fused_moe_triton/configs/E=8,N=8192,device_name=NVIDIA_H200,dtype=fp8_w8a8.json",sha256=Uz5X80VcNBOaxshwVNUEittHk2zqB4HQCfTJ4TPG5aM,3274
268
- sglang/srt/layers/quantization/__init__.py,sha256=UOQcyCvKFkX0u_OPPex7X5X98iUR3lXgBnLbffu0n9g,12424
290
+ sglang/srt/layers/quantization/__init__.py,sha256=WVaItwaovrn-tZiAK0Wvs5RkV_yXi88K4z3xHB44Wf8,12424
269
291
  sglang/srt/layers/quantization/awq.py,sha256=KemDG55U3B6YZVjMV71awVAIj0islFvtxcUHmOBeGy0,6739
270
292
  sglang/srt/layers/quantization/base_config.py,sha256=jWk_egQrVNMYmQgbTI9vkcgzScLFjB5_sywFlAfE5J0,4776
271
293
  sglang/srt/layers/quantization/blockwise_int8.py,sha256=cu9-JiCZDfMfvB97Kv_-eEG87VX5bRFIllFkzpO_xIg,15122
272
- sglang/srt/layers/quantization/deep_gemm.py,sha256=UFzsd0iiqVTBo0Ow_6ylVVFK8B9EUWTNQQYGvsNfm2s,13129
273
- sglang/srt/layers/quantization/fp8.py,sha256=Niu89OfZd4pIpkwZ1zd2Hrlffx0c5L5zkix6Lzi0Vys,40970
294
+ sglang/srt/layers/quantization/deep_gemm.py,sha256=VmSSrhswu_vL0ecmQaBZsBTPYSX5OfjxVBgCGYXLUug,13141
295
+ sglang/srt/layers/quantization/fp8.py,sha256=da-6ji_HBISKwIgMMX-JGlDKMLi-qL9j2XLer5cFAsU,40945
274
296
  sglang/srt/layers/quantization/fp8_kernel.py,sha256=C2_hOLRO27-Yvjy-Nm2niehD2gWSMuP6TnNX07ESqh4,32018
275
- sglang/srt/layers/quantization/fp8_utils.py,sha256=1AmKsAFlpPcwbqpFR6P2CaCOvpp8dYY7lzXLoKjVikE,18631
297
+ sglang/srt/layers/quantization/fp8_utils.py,sha256=vqH-bMb2DD0A7Y7hZjN-TGTg5h6aJ-cLW9H2adyZzqk,18651
276
298
  sglang/srt/layers/quantization/gptq.py,sha256=gyGMOPXHzozK7pPWSjKgLdFX9h7MCEww7n8FqEVEVac,15364
277
299
  sglang/srt/layers/quantization/int8_kernel.py,sha256=CR-VuTTR4GYluOZTpS5mmEz3hYrsY4GOX-G-h3XAYKc,12163
278
300
  sglang/srt/layers/quantization/int8_utils.py,sha256=YK9CS-lb_n91kNCTKK5o5apYF31V2giDg5G5VKrpcUA,2356
@@ -442,10 +464,10 @@ sglang/srt/layers/quantization/compressed_tensors/schemes/compressed_tensors_w8a
442
464
  "sglang/srt/layers/quantization/configs/N=7168,K=256,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=PD4AJYCkHfy2ivv9baMouFXzBTy0eKMumbAfxfm91HI,3256
443
465
  "sglang/srt/layers/quantization/configs/N=7168,K=256,device_name=NVIDIA_H20,dtype=int8_w8a8,block_shape=[128, 128].json",sha256=FImA-TJ_tQDjqwoNWxS--sRDoKDXf9gamlME3tkxH58,3252
444
466
  "sglang/srt/layers/quantization/configs/N=7168,K=256,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=FFBjSWlpKXMxfAUUYUqXbOK_Hd7qBeBsfbcaa9uB4qY,3249
445
- sglang/srt/lora/layers.py,sha256=cu1kqDCuH05ck8HVtwmVuMVBzcPJZeDY3mk486teB4E,11848
467
+ sglang/srt/lora/layers.py,sha256=xdP2Gwlw9PCPZBhujGqO6aBn0eGxpVeIBFUp1LIGCto,13119
446
468
  sglang/srt/lora/lora.py,sha256=uNvbjZ_Wr1SLI9-ElRJA_JKwkibSGroP5Bfpsr9MI-Y,7527
447
469
  sglang/srt/lora/lora_config.py,sha256=qDgMTx_69jyJUl29O5FxLzYa0BMhqYVXWXfyyVOvGm0,1684
448
- sglang/srt/lora/lora_manager.py,sha256=nyqkm7RLoQE6myfqcH9r0zwME4aEZ3pFkVjY36QTlvA,9200
470
+ sglang/srt/lora/lora_manager.py,sha256=u8jwO1IeAb7zE47YU-2fXwx7RmxXaYCc39uJxvYHoHY,11797
449
471
  sglang/srt/lora/mem_pool.py,sha256=xUFoHUDJgX9lt2YugD9HUY5tIMnJiazYMZ6LYqSGv-E,9633
450
472
  sglang/srt/lora/utils.py,sha256=GjEBgsGhDhX4NqVqeaciznQ8RotKZmb2c-nw4YMLHxA,5251
451
473
  sglang/srt/lora/backend/base_backend.py,sha256=EIz8I-GIrdmK4fISw3ENhbJVVITaxKfyLxHXGPU4fPs,5044
@@ -458,26 +480,27 @@ sglang/srt/lora/triton_ops/sgemm_lora_a.py,sha256=ZmWEqHJaorRNNj-c_ZXPi_pX8X_yIA
458
480
  sglang/srt/lora/triton_ops/sgemm_lora_b.py,sha256=Q58UzWUb3QFqY_ZxWA3poN373N0Hwkks5AQRKIuvFC8,4517
459
481
  sglang/srt/managers/cache_controller.py,sha256=d4RGqbut1FlzJnpqr7WY_TYmRjYPS07OoOVbztjs5xI,18959
460
482
  sglang/srt/managers/configure_logging.py,sha256=fOJaXAQ1n9m-8KPJndpsKvS885i69SMafoEADLIVfIM,1633
461
- sglang/srt/managers/data_parallel_controller.py,sha256=Oo-0sbF0W1fcpw88-iKH_7pttYjWl8IHCePcuF3rU5c,10894
483
+ sglang/srt/managers/data_parallel_controller.py,sha256=UgMruoTjQDWDCZK7ATmmgNrrY011pDqrFGl4vJdBpKU,11677
462
484
  sglang/srt/managers/detokenizer_manager.py,sha256=3S3aRvKSi75RQSxEEQkeyxKDNNunWiw9wlwsbT1VXSo,10099
463
485
  sglang/srt/managers/expert_distribution.py,sha256=r3o5RGI0gnV7xb60AApqKYa0oiSB37oB7hQBX7P3xZM,3225
464
486
  sglang/srt/managers/io_struct.py,sha256=9mdBGOkblguT1x6Ds9wL3j0MWAQiUQVdVRL4a7IUnA4,31631
465
487
  sglang/srt/managers/mm_utils.py,sha256=JTu5B7jZWTtZi8LCpVa6ITvSToxcuf5PDbb3FJC9M6o,18089
466
488
  sglang/srt/managers/multimodal_processor.py,sha256=XlRYvNhF6XOssreRX9DZPhLSpps_VE62gSKw3EGdNPo,2088
467
- sglang/srt/managers/schedule_batch.py,sha256=NXeAXZVvXcB8ApByiDLNaFww3aNQl_rZDxQKujhD5uo,63553
489
+ sglang/srt/managers/schedule_batch.py,sha256=E_Q3v4F7oioPqQFu891NlY_UK-9O6dJ7GnitjnKa1LU,64272
468
490
  sglang/srt/managers/schedule_policy.py,sha256=E1qVq2G3jptKdX9nlqfayeRBUll9xB6bK8nBf3EW32E,19469
469
- sglang/srt/managers/scheduler.py,sha256=3EJ__bAck1xF8GWR6hFlt0HU7VwHMO7uM2eXKHOf3KI,80627
470
- sglang/srt/managers/scheduler_output_processor_mixin.py,sha256=u2sj6MViFTov0lVZSysZ-wph2pEqRCtCjwA1UdttZ7I,26338
491
+ sglang/srt/managers/scheduler.py,sha256=OuSUf31bxFSx4tUbNggAX3DyIfD4iuJoLwJbyp5Py64,89816
492
+ sglang/srt/managers/scheduler_output_processor_mixin.py,sha256=15Eicph3bPVuBMPsMPOLReNc2Kmi6m1WXlq0UbYtj9g,26773
471
493
  sglang/srt/managers/session_controller.py,sha256=o-ifit0n4_xHLNmyD0Ams8FxGRgxFybX-Vz1hwgr3UQ,5755
472
494
  sglang/srt/managers/tokenizer_manager.py,sha256=4l4PAvfQrJqlYADQbl7cgpLhBBY52pzI5AzRYIzAjLs,50693
473
- sglang/srt/managers/tp_worker.py,sha256=OV0Zc-oK-HXSWO6x26ItDEf-qB2IqyJq2z0TFuBZz0s,8970
474
- sglang/srt/managers/tp_worker_overlap_thread.py,sha256=3_ZJ8Rq7v2ZDaRNTRu5Dy8AbqiAlJQp3IAKnn_WAwd8,9127
495
+ sglang/srt/managers/tp_worker.py,sha256=zo9OLsBABpBUyBgTKPPcDhtrLkRB_edyN9ANzGz_Pzc,10228
496
+ sglang/srt/managers/tp_worker_overlap_thread.py,sha256=PyBiUdHeh1Z_o_R34lNB28SBjqTP4nArNCQhX0O6K2M,9440
475
497
  sglang/srt/managers/utils.py,sha256=5i75uLlQOF_5CaT02CrWtwozMTtwTg2_nLP8Dtr-JZQ,1536
476
498
  sglang/srt/managers/multimodal_processors/base_processor.py,sha256=ata9H6Ry4QfqBoA_g0auG0sMnKfGrlZn74lM77ihtiA,10172
477
499
  sglang/srt/managers/multimodal_processors/clip.py,sha256=lRc2mcuDbAhZVf-0EfkO81pqDiol9zLvTpDqtPIBQ2k,1525
478
500
  sglang/srt/managers/multimodal_processors/deepseek_vl_v2.py,sha256=hpjpGFzlRBQ8Xv08i37X_VUhnDp_Qm2xD1_F17vK8fI,3253
479
501
  sglang/srt/managers/multimodal_processors/gemma3.py,sha256=G52ck_3UQGeyrtvjLqI8B0Tm8iNsyB_ahiMTAvx083U,2191
480
502
  sglang/srt/managers/multimodal_processors/janus_pro.py,sha256=UJoKQWsoU9kittKDwjWbG2KC12wSA-4A3DpTPhA6VoI,1854
503
+ sglang/srt/managers/multimodal_processors/kimi_vl.py,sha256=vC9OeS7gVTHzazbluiQ1I0QRKqszlqK75ghUA1rmUNc,2490
481
504
  sglang/srt/managers/multimodal_processors/llava.py,sha256=8mac3vUUpVd12o43k1TyMaLEySZB915ks8Q5epeZmbg,6209
482
505
  sglang/srt/managers/multimodal_processors/minicpm.py,sha256=uEnlsImjHBOMVNGlfBGpn1zCDLNeMY58HvJ7ZthL2N4,5698
483
506
  sglang/srt/managers/multimodal_processors/mlama.py,sha256=MLiGS606LzVtdoXvjWGANx-K_7nE9J_fMVmkXN7Gz8k,1661
@@ -487,14 +510,14 @@ sglang/srt/mem_cache/base_prefix_cache.py,sha256=NY62Zo0A0tLJ7ObRLOQqQcXCxoJUDZs
487
510
  sglang/srt/mem_cache/chunk_cache.py,sha256=it5SfL1FwMbrdeOH-I-Eu_i-I9hFB1xL-z_brIUoCkk,1835
488
511
  sglang/srt/mem_cache/flush_cache.py,sha256=GYcxmNXh4hsMpFfNOuCTpKilW7guZwTtAg_usVeM3J0,979
489
512
  sglang/srt/mem_cache/hiradix_cache.py,sha256=BJR-R2u5YyYIhGIxTY-3rf8Vx60XjCRU8Yhmkn2fzzM,16597
490
- sglang/srt/mem_cache/memory_pool.py,sha256=J2eAAefAl0TIejH7h-hwz_ak_T-fSh_e45tUNrhX0BE,34599
513
+ sglang/srt/mem_cache/memory_pool.py,sha256=ZJYAceAtEv326tYERXsHeBy7_0HGnk_jpfEKJTJwPMU,36260
491
514
  sglang/srt/mem_cache/paged_allocator.py,sha256=BrJS0vN1k-vTSgb_M8u_1KoZFRgzgR1WRyImCTq3T0U,9770
492
515
  sglang/srt/mem_cache/radix_cache.py,sha256=Lm-pco6CJ4orb9IfDpbHm5MnyK8Ya0OF1x9p88dv548,14906
493
516
  sglang/srt/metrics/collector.py,sha256=zHg4twFQJvuK1mSme3-EYQa9PJryfp_u7a4RxQ5RcO0,8874
494
517
  sglang/srt/metrics/func_timer.py,sha256=VFyNRrbnKVCwnQsrlLin1lITJfjQpf9m8sGPqL5LIsQ,3438
495
- sglang/srt/model_executor/cuda_graph_runner.py,sha256=iFryO9dglpnFCoNWxZqKdUhQycT8In29C0kIba3G1Dw,23687
496
- sglang/srt/model_executor/forward_batch_info.py,sha256=T9B5vWaJwlKUH0fQTPe3XdbkTYEUI6iKxBxUHs-cAMM,26632
497
- sglang/srt/model_executor/model_runner.py,sha256=Eei9JGCX-SQaTF1T4iBS0zg4UPdyjQ67UkKwgbK_200,46136
518
+ sglang/srt/model_executor/cuda_graph_runner.py,sha256=ISDLqpJZ0_WjX2IqWt6yASy4yLMUchVc9-6J_bK-UBY,26208
519
+ sglang/srt/model_executor/forward_batch_info.py,sha256=Kz30RuEjuOAN9_8hlvvknF4qeohyas7NrS90FCRtIMg,27730
520
+ sglang/srt/model_executor/model_runner.py,sha256=FYNlATMIQ-zC_NDqpWu3hBu7A-rxhgaPqF029U7txvo,48596
498
521
  sglang/srt/model_loader/__init__.py,sha256=zGZkOBz1zx-pkaIy47BasL3fjDlAcxAXUTjInOhXHAE,919
499
522
  sglang/srt/model_loader/loader.py,sha256=YYmtvkQw0B1qgPw0_gN-K4yy7CEYbTSR__0Dl1Fnm6k,55342
500
523
  sglang/srt/model_loader/utils.py,sha256=0NaMR67fESFopaklmsleiL27XH1QUrjZW246MUu1EJ0,1369
@@ -507,8 +530,8 @@ sglang/srt/models/commandr.py,sha256=5Y_b3K0QY7D37nFGkyiGgY38RleRui_GJUYcHSuHUZo
507
530
  sglang/srt/models/dbrx.py,sha256=4pn_fdoATg01VEqNnIAxNEsKV5XU7gwHyd289eydq1s,15598
508
531
  sglang/srt/models/deepseek.py,sha256=ZnN02HdgXCB23Vno5V9UMUoOxH5HC82vNTwsVulUJ-o,17206
509
532
  sglang/srt/models/deepseek_janus_pro.py,sha256=8wAzvcGdyo--3faMN4QtagT1eAZMhMFduvpCXqUS48Q,70456
510
- sglang/srt/models/deepseek_nextn.py,sha256=XW0PJAvUVx5i1F6liNMooopj833qyQ4Y4ujn3iJDDak,17825
511
- sglang/srt/models/deepseek_v2.py,sha256=6fEihiaHcl9tjawa1GnCKGIappuLnDfmmVChhPswSIU,71820
533
+ sglang/srt/models/deepseek_nextn.py,sha256=ptShwRSOLQbFRU1lJ_uHL1byA8z9d-YKAH4eEiLXPeM,6249
534
+ sglang/srt/models/deepseek_v2.py,sha256=1RJvBzWWTOYJ1U7s65xK1V15QFvJv7mfkNTIYgape-w,73943
512
535
  sglang/srt/models/deepseek_vl2.py,sha256=j8BdxZsMjm6lPdbDipEIKhVIVywCP1Vl1Kl46BZ5_0Y,13147
513
536
  sglang/srt/models/exaone.py,sha256=rX7J0xFt9TSt6tMIhnYMkb5KDnqTJIV4BtjPLFwQ8_8,13425
514
537
  sglang/srt/models/gemma.py,sha256=4cdrPISg1VKnsuI-QPTpYvet4BrX8BMKvCIN82iLskw,12641
@@ -522,11 +545,13 @@ sglang/srt/models/granite.py,sha256=5WOJyNYAlt5RNHSexNfPNihhSxIMd7wPzju1cTixKig,
522
545
  sglang/srt/models/grok.py,sha256=vESZeGS4adI_JAerXIkCcTm15-CNiGeS7VHc36C6w1A,28033
523
546
  sglang/srt/models/internlm2.py,sha256=RDAT9drjdgVEFmCMq99RTn3weMQFhl1NHhkhyDX8f7M,13056
524
547
  sglang/srt/models/internlm2_reward.py,sha256=ndfGmyqYZbVZ7C7rJ-v9oK3wa-EpoBGybS8MlyKZi2E,2522
525
- sglang/srt/models/llama.py,sha256=Y4ROe8ohP84G4vin_Sr_vjG0XRoM5gGgnrojxOXn_uc,24942
526
- sglang/srt/models/llama4.py,sha256=idwV_rEJ_tPMD1iLQzvaJqmux-Osoa3bc0g04VYgu1w,17867
548
+ sglang/srt/models/kimi_vl.py,sha256=YoM6CmrF4ZS0SsXKiR-_SfylKhd87ciJjno6_x5LG4o,12874
549
+ sglang/srt/models/kimi_vl_moonvit.py,sha256=M5L7AJOJ2Zh2pqLZAq8aRqhbTSlipr8XOoD3ix6o2sU,23908
550
+ sglang/srt/models/llama.py,sha256=45bs_qVp53RmnU-zXaHrj2vvhxvmBsWgEemnBtvI8_k,26978
551
+ sglang/srt/models/llama4.py,sha256=xopptG6PXpnGKCMX-o4QAfiWpDpyLUicTPsQKm1rorI,17942
527
552
  sglang/srt/models/llama_classification.py,sha256=4QWTFaUZIFKYZvEzs8bx8VkOZNIwdYCLrnwrdAw4QK0,3108
528
- sglang/srt/models/llama_eagle.py,sha256=OB2lKsjn7BcfCZljklnhk83me8j0PuQmYLou7baNcq4,4866
529
- sglang/srt/models/llama_eagle3.py,sha256=v3bftBVDIGjnzngQYnu19cy0J_3w7yruHqLP5nsAQDM,6642
553
+ sglang/srt/models/llama_eagle.py,sha256=Ubh_jmtfiOnriwvHgQT0ZGID7JvYvdSi_QGHOIkEgPE,5028
554
+ sglang/srt/models/llama_eagle3.py,sha256=KCvUrWn17t7T28G60HwMyy99iB9AbdbPnS2el9n3r6o,6804
530
555
  sglang/srt/models/llama_embedding.py,sha256=zq-_lNu35VBFc7eemiam0zdkGIE8fzrgk5OWYfirZnA,3254
531
556
  sglang/srt/models/llama_reward.py,sha256=LF2nqMV5XOrljGjAwJg43mBv3z6Q040I2EYlgZeCp8k,4681
532
557
  sglang/srt/models/llava.py,sha256=KMwNNrlMuMaKEOZMDRBKBQbe6uctpKTLc0zOceyGC34,27242
@@ -549,11 +574,11 @@ sglang/srt/models/qwen2.py,sha256=ab912Yyk0aXOzI-wrxqN-sNF3bTVkNxB8P2uNcOCv9U,16
549
574
  sglang/srt/models/qwen2_5_vl.py,sha256=mqgmDHleJN3GmuZG1pUSpIZYKO1omTsa8P5MXYULAGM,22462
550
575
  sglang/srt/models/qwen2_classification.py,sha256=dGrMm4ebd30_lBhHOhaV57ig2iOTx3nqB4GEzsrRIM8,2747
551
576
  sglang/srt/models/qwen2_eagle.py,sha256=Iz0HWL2FgSD3FqoFhfYmbIZeEYkPTJ96lYbkncmHJX4,4644
552
- sglang/srt/models/qwen2_moe.py,sha256=bmS2pyHD5zQo5plTCzAo_mjnahVtJ1jaRSURX1PlQC4,18313
577
+ sglang/srt/models/qwen2_moe.py,sha256=iG0d2WwUosOFR9w2YGM9CfvZ1NG-rhse3OUTwSs8a6Q,18567
553
578
  sglang/srt/models/qwen2_rm.py,sha256=-mQXDEv11p-I1HXgYLTtY6ROem6UYorO958WsDrzsgs,2837
554
- sglang/srt/models/qwen2_vl.py,sha256=tgES87Rmdl7hqMLAnxYqpWerxK28n5UY7Ma5309TPqs,21408
579
+ sglang/srt/models/qwen2_vl.py,sha256=EaDBXbzeIToYUUzs0yyByXE66-G1BCfATXJgkaHa-PE,20975
555
580
  sglang/srt/models/qwen3.py,sha256=reaowGkotYAGHS5zTCWrvnyxtu92QKus19n-2amtMa4,12358
556
- sglang/srt/models/qwen3_moe.py,sha256=Tee7oW6Xvo2pV_Q93y-HKykBFiPjo_-YfeIsIelB3hA,15623
581
+ sglang/srt/models/qwen3_moe.py,sha256=A9Z3OhJqld1sJUDsHymgGxib4lMCMTKxF8iIzHDGdNo,15877
557
582
  sglang/srt/models/registry.py,sha256=inKh9iwOp3LFYm3nqujg-OtABClOP-ifc1stA9cZegA,3434
558
583
  sglang/srt/models/roberta.py,sha256=Zgd35och3pW6TYrNeEoeOZ8qPfbFwB3ngThpVWSPBcY,6320
559
584
  sglang/srt/models/stablelm.py,sha256=0x_31uIr3WcWwecdPAI3ek9KkyKBJS7VwknTk2y0gjY,12281
@@ -561,8 +586,8 @@ sglang/srt/models/torch_native_llama.py,sha256=5tfFSMAXB3ScToqTALtCXa8Oo-qPCJh-K
561
586
  sglang/srt/models/xverse.py,sha256=DsNVI9JpzN4jj0Ry6aTrj7r-xq5YLOoDX2kH4YLJA-I,14035
562
587
  sglang/srt/models/xverse_moe.py,sha256=7KCM2-j12towDMNvXkuuYiBOmNauH6NG4Ip40x0khqA,16782
563
588
  sglang/srt/models/yivl.py,sha256=oToK7-u5IGO7xwpJIQ7VtudlK6-zPqJX4bt6_wv0SH8,4850
564
- sglang/srt/openai_api/adapter.py,sha256=MMHLsz6Thepwtr4rtOlYwOlaKk212vgpC0p5nmqGt5U,75550
565
- sglang/srt/openai_api/protocol.py,sha256=1IVfYt0LUNLqFz0DNZyqxU02IRYVHcL7jnpMG9bNq34,13532
589
+ sglang/srt/openai_api/adapter.py,sha256=mMaaN2hVhRARIzfXFfmqORYLO8bXUi-GCOJLZq0lYFg,77283
590
+ sglang/srt/openai_api/protocol.py,sha256=eTSp9x4uFABDxTYRnnftgoVSF5PWlkixnFTfQRyG0dM,14142
566
591
  sglang/srt/platforms/interface.py,sha256=hym3iooBB4C8if5hDZezgVN6h4NIOu7sg2ZUBIV6XmM,11246
567
592
  sglang/srt/sampling/custom_logit_processor.py,sha256=tDvoLgLqn-sy1qcY6vSrpbnHCeqbdk0uhMOO-uy4p4E,1099
568
593
  sglang/srt/sampling/sampling_batch_info.py,sha256=4LCowU2bk0TOSfIGpEy90N1SpTsiOKK8Rx1ZYcklUFQ,11988
@@ -575,14 +600,14 @@ sglang/srt/sampling/penaltylib/presence_penalty.py,sha256=NRh10AJrrQlGJ6S-enGdRe
575
600
  sglang/srt/speculative/build_eagle_tree.py,sha256=lt4sXUehPi26MT2-2Z0VivtF6AP7kirSaEO_u-YJ4J4,11670
576
601
  sglang/srt/speculative/eagle_draft_cuda_graph_runner.py,sha256=NviXdUvowQkV1kLs3eXLlxJx6UZzyQMZH03zCXpsIg4,9291
577
602
  sglang/srt/speculative/eagle_utils.py,sha256=iJYhklXHfDgEKbVB39HkVEea-XTEC60Z_LjIVjkrZQs,28701
578
- sglang/srt/speculative/eagle_worker.py,sha256=D4G8hnwtc8xQt1okG4TY9wYSXbKTqGVDAD22AUXW6pA,26824
603
+ sglang/srt/speculative/eagle_worker.py,sha256=MwsBbKyV-dCwzYlIpVcb-urk-GSdoe_kY8KHe5Gkw7A,26860
579
604
  sglang/srt/speculative/spec_info.py,sha256=rhaKG0TzyF9XZYHEWp1jccwTBohSNsUDvxHFtAoOl18,709
580
605
  sglang/test/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
581
606
  sglang/test/few_shot_gsm8k.py,sha256=7VLbWl4nCQs1wjtW4q-46jf9jUCycSs5Iw8v7sUSzBw,4284
582
607
  sglang/test/few_shot_gsm8k_engine.py,sha256=QQbrwOX6-cJDD3RZC_e7zPnt6aSo8JdF8X_lRHSjdDM,3886
583
608
  sglang/test/run_eval.py,sha256=9yO0hXZOcn4abEOs96T-XPguDEklK16Ltco0pGF3zCg,4020
584
- sglang/test/runners.py,sha256=vSOl38rVDR3l2ezVCs672vE-LcOA2rJHjlkhLgEjcz8,30260
585
- sglang/test/send_one.py,sha256=6FhbJ3c8RpXxvFTELRXaF97GpT7zXXsCDYZh1DqG22E,2550
609
+ sglang/test/runners.py,sha256=WWAu07NXSJV1y4W-iEi_iOCy1P5Ow9rL0ex-U969Nws,30417
610
+ sglang/test/send_one.py,sha256=_l72sRfuXRUldyD3PD63hg_WxNvvhW5unNnbe4XuAwk,4380
586
611
  sglang/test/simple_eval_common.py,sha256=joqrGysuLnJFtzDRIgFkMsRyKUSyjVPFWp0_PHAL3Ik,12378
587
612
  sglang/test/simple_eval_gpqa.py,sha256=8Xt9Bw05c7SZTYrCZgB68OZUqUbLo69ywiyx0bTvSUk,3220
588
613
  sglang/test/simple_eval_humaneval.py,sha256=zmV3xWYc2OrpiT9Dy55RTKZL5DEROD1cJ0NA_-cU5zI,5685
@@ -596,13 +621,13 @@ sglang/test/test_custom_ops.py,sha256=2bSo9P5_rJZYFq8Y8IKRimDfFyZZGJluhL7Ngny0Pf
596
621
  sglang/test/test_dynamic_grad_mode.py,sha256=L76yUCuk_ymNpXD2CmO8r2GiGjIvD_gtTsuFDs2NolI,1638
597
622
  sglang/test/test_layernorm.py,sha256=2GMWqqNDuGvSMSsEBF5eDCzwVSYA9E6hGhRo6s4ecKg,3764
598
623
  sglang/test/test_programs.py,sha256=VZ3vXtUDBnXz0M7gFdDH8hXg9Wa0j_qI8CVqjEgRN_E,18877
599
- sglang/test/test_utils.py,sha256=OPTl_PKVlHitrRJW5DFClpQA6xN6FvxatiOpgngCuJU,31153
624
+ sglang/test/test_utils.py,sha256=WqLvYA2N18USdHWY2kKmrIixLtiOb3r4HJ2uT1kRu1o,32905
600
625
  sglang/test/attention/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
601
626
  sglang/test/attention/test_flashattn_backend.py,sha256=_rTG849FwQdVTyGKkqhczaOqngBmRWXFmkl5NnuK1GM,13914
602
627
  sglang/test/attention/test_flashattn_mla_backend.py,sha256=g4O50WblTpM7_Gq2b76k0i25_z01BOUBQ4i6PmyxpO4,10774
603
628
  sglang/test/attention/test_prefix_chunk_info.py,sha256=er0i3KGHMkw-4UZB1GCFd4oYwRcXfU5wpO1ORqpNGGA,7626
604
- sglang-0.4.6.dist-info/licenses/LICENSE,sha256=FJXh51fvTQklojUFY89XVLsjxRcBqOxPs8XNy-2uZ0c,11346
605
- sglang-0.4.6.dist-info/METADATA,sha256=ITL_ELbps-k7hM4JnrS_YQCgwSO1glI5-dE9kX0Hft0,25361
606
- sglang-0.4.6.dist-info/WHEEL,sha256=ck4Vq1_RXyvS4Jt6SI0Vz6fyVs4GWg7AINwpsaGEgPE,91
607
- sglang-0.4.6.dist-info/top_level.txt,sha256=yxhh3pYQkcnA7v3Bg889C2jZhvtJdEincysO7PEB09M,7
608
- sglang-0.4.6.dist-info/RECORD,,
629
+ sglang-0.4.6.post2.dist-info/licenses/LICENSE,sha256=FJXh51fvTQklojUFY89XVLsjxRcBqOxPs8XNy-2uZ0c,11346
630
+ sglang-0.4.6.post2.dist-info/METADATA,sha256=XZjs979dHuGsAVGlpT-ZQd9msQWxI4mN1WTtmG6CPvo,25419
631
+ sglang-0.4.6.post2.dist-info/WHEEL,sha256=wXxTzcEDnjrTwFYjLPcsW_7_XihufBwmpiBeiXNBGEA,91
632
+ sglang-0.4.6.post2.dist-info/top_level.txt,sha256=yxhh3pYQkcnA7v3Bg889C2jZhvtJdEincysO7PEB09M,7
633
+ sglang-0.4.6.post2.dist-info/RECORD,,
@@ -1,5 +1,5 @@
1
1
  Wheel-Version: 1.0
2
- Generator: setuptools (80.0.0)
2
+ Generator: setuptools (80.1.0)
3
3
  Root-Is-Purelib: true
4
4
  Tag: py3-none-any
5
5